Fatih Ozorpak

Mountain View, CA

650-***-****

ad29xx@r.postjobfree.com

www.linkedin.com/in/fatih-ozorpak

Data scientist

Professional Summary:

Having 9 years of IT experience in performing data driven analysis as AI Engineer and Data Scientist product-based projects (project 2 details in below) and 5 years in Data Analytics in computer science and physics.

Data scientist with a strong background in machine learning and a passion for turning data into actionable insights.

Experienced in product ownership and end-to-end model management,

Experienced data scientist with a passion for research, technology, and innovation. Leveraging a strong foundation in contemporary machine-learning techniques.

I actively explore emerging technologies to drive data-driven modeling and optimization of complex systems.

In business problems directly working with clients and returning to our model and tuning it with client’s specific highlights and tuning our model in the light of statistics solving the problems also applying machine learning algorithms forecasting predictive analysis for solutions, dimensional data from varying resources.

Managed budgetary processes and financial planning for herkul education to ensure effective allocation of resources.

Successfully implemented cost-saving initiatives, contributing to a 20% reduction in expenses within the first 6 months

Collaborated with cross-functional teams to monitor and analyze financial performance, providing regular reports to senior management.

Performing decoding, data cleaning, data mining, data visualization, statistical programming with activities using Python,

Prompt engineered and fine-tuned LLMs using OpenAI, Bert classification

Collaborated with cross-functional teams

Technical skills:

Programming : Python, SQL

Machine learning: Linear Regression, Logistic regression, Decision Tree, Random Forest, SVR,KNN, PCA, Bayesian Regression, Artificial Intelligence, Naïve Bayes, Time Series, Bert training models

Data Visualization: Matplot- lib, seaborn,Plotly

Libraries: NumPy, SciPy, Pandas, Scikit-learn, Spark (pyspark), Seaborn, Matplotlib, Statsmodel, TensorFlow, Keras, NLTK, LLM, Bert

Database: MySQL, PostgreSQL, Hive, Oracle

Data Analysis : Teradata, BeatifulSoup4, Power BI, Tableau

Cloud Computing Tools: AWS (Domino Datalabs, Databricks, Amazon Redshift),Data validation,NLP ( natural language processing), Version control software (Git), Optimization tools.

Education

M.S. Biomedical Engineering, Bogazici University, Turkey – www.boun.edu.tr

B.S. Physics, Bogazici University, Turkey – www.boun.edu.tr

Languages: English, Turkish, Portuguese

Experience

MegaLabs Palo Alto, CA

Project: Text Document Cleaning and Sentiment Analysis of Twitter Data

• Led a comprehensive text document cleaning project, extracting valuable insights from Twitter data.

• Implemented a robust labeling process using Chat GPT prompt engineering, ensuring the effectiveness of subsequent machine learning tasks.

• Employed count vectorization techniques to preprocess labeled texts, facilitating feature extraction for machine learning model training.

• Conducted machine learning experiments with various algorithms, ultimately identifying Naive Bayes as the optimal choice based on accuracy metrics.

• Introduced advanced NLP techniques, including fine-tuning with Large Language Model (LLM) BERT classification, resulting in a remarkable improvement of over 20% in accuracy.

• Orchestrated the evaluation of BERT-based models, showcasing superior performance compared to alternative methods.

• Successfully navigated the intersection of data preprocessing, machine learning, and advanced natural language processing, achieving robust and accurate sentiment analysis results.

Data Scientist/AI Engineer Mega Labs .(Palo Alto-Ca) Jan 2020 – Present

Responsibilities:

Implemented data models, algorithms to develop machine learning solutions as a member of data scientist team, LLM.

Collaborated with product management teams to understand business requirements and translate them into data project.

Led a cross-functional team to develop and deploy a customer churn prediction model, resulting in a 10% reduction in churn rate.

Conducted prompt engineering experiments and fine-tuned LLMs (ChatGpt, Bert) to generate AI ad-creative variations.

Conducted regular model performance evaluations, fine-tuning hyper-parameters for optimal performance.

Organized and conducted machine learning pipeline, with a focus on data preparation, model training and optimization,

Successfully used visualization techniques to inform and give intuitions to stakeholders,

Discovered patterns, formulated, and assessed hypotheses, translated results into strategies which drive growth resulting in increased revenues and improved customer satisfaction.

data access and analysis methods for large customer and transactional data, with expert SQL skills and working experience in data extraction, manipulation, and summarization languages and toolsets, including, Python, Databricks, Azure.

Developed post-processing and filtration pipelines for LLM with AI generation Process.

AI experience using visualization and dashboard tools, such as Tableau and PowerBI

Strong business acumen, complemented by superior written, verbal, and presentation skills.

Proactive in anticipating operational and program issues and developing preventative measures.

Applies fully functional professional knowledge in Analytics.

Project2: Service Analysis and Churn (Mega Labs Palo Alto) 2021-2022

(shared with dream venture) Prediction / Industry: client churn reporting

Responsibilities:

Feature engineered, handled missing values, and created new features by combining existing ones,

Utilized Python matplotlib and seaborn libraries for visualization,

Applied ANOVA, chi-square statistical tests to determine the predictive power and association amid the features,

Selected the most significant features to decrease complexity and processing time and increase robustness of machine learning models,

Created different models including Regression, Random Forest, SVM, and Neural Networks algorithms; compared model performances, and adjusted hyper-parameters to obtain improved outcomes,

Delivered a probabilistic churn prediction model, achieving 4% improvement in F1 score over existing model.

Developed a Tableau dashboard to visualize the churn prediction results and the features that influence the churn and Data Validation for the new costumers.

Project1: Detection of Sarcopenia-related morbidity and mortality.

Responsibilities:

Diagnose Sarcopenia among aging people, depends on early detection and treatment, more importantly on unraveling the relationships between muscle loss, decline in muscle function and physical performance and the aforementioned potential are predictor parameters for a further modeling.

Explored the factors that influence of factors for Sarcopenia in men and women separately.

Analyzed and removed non-informative and redundant predictors using supervised and unsupervised feature selection metrics for better model interpretability.

Determined the probability that the patient has Sarcopenia based on variables available when the patient is home, such as age, weight, height, exercise, Hypertension, DM, Education, Smoking.

Delivered a two-stage probabilistic model, optimized for F1 score and recall, to predict Sarcopenia in patients as strong positive, positive, and negative; and an accompanying dashboard to visualize and inspect the results and drive further analysis.

Increased classification metrics over baseline as 3.5% increase in accuracy and 2% increase in F1 score.

Administração De Bens Ltda- Data Analyst (Rio Grande do Sul/Brazil) 08/2018-12/2019

Python and sql analytics for ETL

Performed market research for ag-tech startup; weekly C-level presentations for 5 major investments

Data analysis and visualization of asset management and potential investment opportunities for a $3M+ fund

Reporting the results with Tableau, dashboard and stories

New investment area and how to cut some extra fees, in international investments

Central bank and financial investment controls and reports

IT Manager -Data Analyst Colegio Belo Futuro 2013-2018 (Sao Paulo-Brazil)

Led the IT department at an international college, overseeing a team of professionals.

Managed and organized IT lessons for students, ensuring a seamless learning experience and technical proficiency.

Utilized student analytics to assess technology needs, optimize learning processes, and enhance student engagement.

Conducted financial analytics to track and optimize IT expenditures, resulting in

Held regular meetings with stakeholders to align IT strategies with organizational goals, fostering collaboration and ensuring effective communication.

Successfully implemented Chromebook integration and online system financial and educational system for students, parents and school administrators.

Contact this candidate